ATTESTED vs. VERIFIED vs. AUDITED: How Proof Tiers Work

Not every claim of "we have users" or "we have a customer" carries the same weight, even when the underlying number is identical. A proof tier system labels how a number is known, not just what the number is - three tiers, each answering the question "if I asked how you know that's true, what's the honest answer?"

ATTESTED - the founder's own word

The founder reports the number themselves, with no independent confirmation attached. "We have 40 active users" with nothing backing it up beyond the founder saying so.

This tier is not disqualifying - it's still real information, and it's the fastest, cheapest way to report a number. The point of labeling it ATTESTED isn't to discount it, it's to be honest that the only source is the person with the most incentive to round up.

VERIFIED - signature-confirmed

The claim has a paper trail that exists independently of the founder's own reporting - a Stripe-confirmed payment, a signed agreement, something a third party can check without needing to trust the founder's account of it. (How the strongest version works - the processor signs the event, the machine verifies it server-side.)

VERIFIED numbers carry more weight than ATTESTED ones precisely because they don't depend on the founder being accurate or generous in their own favor. A VERIFIED customer count is a number someone else could, in principle, go check.

AUDITED - confirmed by the people involved

The people actually counted - the users, the pilot customers, the cohort members - confirm it themselves, directly, typically through a simple one-click process. Critically, an audited count records the answer whether or not it flatters the founder: someone who says "no, that's not quite right" still gets counted honestly, because a system that only records favorable audit responses isn't actually auditing anything.

This is the rarest and hardest-to-fake tier, because it's the only one where an unflattering answer stays visible instead of getting quietly filtered out.

Why the tier matters more than the number

"12 VERIFIED users" is a stronger claim than "500 users" with no tier attached at all - the second number could mean anything from a Stripe-confirmed customer list to a founder's optimistic guess, and there's no way to tell which from the number alone. A tier system fixes this by making the claim's own confidence level visible instead of implicit.

Why this shows up as a structural rule, not just a norm

On launchduels, evidence tiers aren't a suggestion for founders to follow voluntarily - every number on the board is required to carry its tier, publicly (R-22), because a proof record that can't say how it knows isn't proof. The alternative - a leaderboard of unlabeled numbers - rewards whoever's most comfortable rounding up, which defeats the entire purpose of a proof-based system in the first place.

The practical habit, even without a formal system

You don't need a platform to start doing this. Next time you post a metric - in a pitch deck, a tweet, an investor update - attach one honest clause about where it came from: "(self-reported)," "(Stripe-confirmed)," "(user-verified)." It costs a few words and a little bit of shine. It buys a number people can actually decide how much to trust, instead of one they have to take entirely on faith or entirely discount. (What that habit looks like in a build-in-public feed.)

The evidence-tier ladder is how the indie board counts - every number, its tier, in public. See the open chairs →